playlists
titles for each score
genres
summary scores
# A tibble: 1 x 12
mean_speechiness mean_acousticne… mean_liveness sd_speechiness sd_acousticness
<dbl> <dbl> <dbl> <dbl> <dbl>
1 0.0585 0.659 0.123 0.0437 0.302
# … with 7 more variables: sd_liveness <dbl>, median_speechiness <dbl>,
# median_acousticness <dbl>, median_liveness <dbl>, mad_speechiness <dbl>,
# mad_acousticness <dbl>, mad_liveness <dbl>artist features
summary ludwig
mean_speechiness mean_acousticness mean_liveness sd_speechiness
1 0.05950926 0.5870416 0.1396762 0.04655653
sd_acousticness sd_liveness median_speechiness median_acousticness
1 0.2967321 0.102002 0.045 0.6205
median_liveness mad_speechiness mad_acousticness mad_liveness
1 0.109 0.01126776 0.385476 0.0326172Analysis
Introduction
first plot
Hi there!
For this reasearch I am going to look at film composer and music producer Ludwig Ludwig Göransson. In this chart you can see somehow what changes he went trough, from 2013 until 2020. I want to analyse some of his most popular tracks. What made them popular and is there a relation between different popular songs? As you can see Black Panther’s has some really interesting findings. For example, “Kilmonger challenge” has one of the highest loudness (as seen in the point size), and “kilmonger” has one of the highest instrumentalness value. Another interesting finding is the song “Rainy night in Talin” from Tenet. This song has a really high acousticness, high energy and a BPM of 130.
// notitie
Ik heb het idee van mijn portoflio uitgewerkt in een word documment, maar nog niet in een duidelijke lijn. Ik had in gedachte om nog eventueel het nummer “The Mandalorian” toe te voegen omdat dit zijn meest populaire nummer is. Verder is het duidelijk te zien (op de volgende storyboard) dat er verschillen zijn tussen verschillende genre’s. Ik weet alleen nog niet of ik hier verder op in wil gaan, en hoe ik dit dan kan aantonen in de individuele track analyses. Ik denk dat het voor mijn onderzoeksvraag het best is als ik mij richt op de poplariteit en bevindingen in deze chart.
Sci-fi or action?
Genres
Chroma
Killmonger
TENET
Truth
Prediction Indie Party Indie Pop Indie Workout
Indie Party 8 4 5
Indie Pop 4 7 10
Indie Workout 8 9 5
# A tibble: 3 x 3
class precision recall
<fct> <dbl> <dbl>
1 Indie Party 0.471 0.4
2 Indie Pop 0.333 0.35
3 Indie Workout 0.227 0.25# A tibble: 3 x 3
class precision recall
<fct> <dbl> <dbl>
1 Indie Party 0.25 0.15
2 Indie Pop 0.292 0.35
3 Indie Workout 0.167 0.2
Call:
C5.0.default(x = x, y = y, trials = 1, control = C50::C5.0Control(minCases =
2, sample = 0))
C5.0 [Release 2.07 GPL Edition] Wed Mar 17 15:30:21 2021
-------------------------------
Class specified by attribute `outcome'
Read 60 cases (35 attributes) from undefined.data
Decision tree:
A#\|Bb <= -1.259173: Indie Pop (6)
A#\|Bb > -1.259173:
:...duration <= -0.6008952:
:...c03 <= -0.5632606: Indie Pop (4)
: c03 > -0.5632606: Indie Party (10/1)
duration > -0.6008952:
:...energy <= 0.1583392:
:...liveness > 0.8180664: Indie Party (2)
: liveness <= 0.8180664:
: :...c07 <= -1.007623: Indie Pop (5)
: c07 > -1.007623:
: :...c10 <= -1.069226: Indie Pop (3)
: c10 > -1.069226: Indie Workout (11/2)
energy > 0.1583392:
:...c09 <= -2.337745: Indie Party (2)
c09 > -2.337745:
:...c10 <= 0.3039758: Indie Workout (8)
c10 > 0.3039758:
:...danceability <= -0.3527262: Indie Party (5)
danceability > -0.3527262: Indie Workout (4/1)
Evaluation on training data (60 cases):
Decision Tree
----------------
Size Errors
11 4( 6.7%) <<
(a) (b) (c) <-classified as
---- ---- ----
18 2 (a): class Indie Party
1 18 1 (b): class Indie Pop
20 (c): class Indie Workout
Attribute usage:
100.00% A#\|Bb
90.00% duration
66.67% energy
51.67% c10
35.00% liveness
31.67% c07
31.67% c09
23.33% c03
15.00% danceability
Time: 0.0 secs# A tibble: 3 x 3
class precision recall
<fct> <dbl> <dbl>
1 Indie Party 0.381 0.4
2 Indie Pop 0.611 0.55
3 Indie Workout 0.286 0.3 
Breakout 3
